HNHacker News
TopNewBestAskShowJobs

CodesInChaos

4,151 karma · joined December 7, 2012

submissionscomments
CodesInChaos··on Anthropic S-1
Just like the old userfriendly.org comic, where marketing promises nine fives instead of five nines.

https://web.archive.org/web/20220226060634/http://ars.userfr...

https://web.archive.org/web/20220226061736/http://ars.userfr...

CodesInChaos··on OpenAI and Anthropic cut their LLM prices by up to 80% within a few months. Why?
Claude "team" plans are limited to 150 seats per company. So big companies have no choice.
CodesInChaos··on Show HN: Proc-CLI – An agent-friendly CLI tool for process and port management
Why do you use commas to separate multiple targets, instead of multiple arguments?
CodesInChaos··on Show HN: Proc-CLI – An agent-friendly CLI tool for process and port management
Claude is surprisingly bad at killing processes. Had several "Whoops, my process killer matched itself and died before its victims" notifications already.
CodesInChaos··on Overgrown, when plants invade your abandoned tabs
Get the "One Tab" extension. It closes all tabs, but saves a list, so it gets over the psychological hurdle of losing the open tabs.
CodesInChaos··on Roku developer repositories moved from GitHub to GitLab
> I always found it weird that people flocked to github.

Same reason we ended up with git. Their predecessors were significantly worse, and then they had enough momentum to keep going.

CodesInChaos··on Roku developer repositories moved from GitHub to GitLab
Downtime aside, what do prefer about GitLab?
CodesInChaos··on PHP 8.2.35
> one of the main performance tricks for CGI was to get processes to stay alive

How does that work? Exec(ve) without fork? That should still clean up the PHP state.

CodesInChaos··on Evolving Threat: http2shell is an RCE in Node, Python, and more
> zstd has high quality rust implementations

On the decompression side yes, on the compression side the first good zstd rust port was announced yesterday. Until then, brotli was better supported in Rust than zstd.

But I agree, decompression libraries should be (mostly) memory safe nowadays, since that side is both easier to implement and more likely to be vulnerable.

CodesInChaos··on OpenSSL 4.1.0
It's just as likely it would end up as an overengineered mess. And that's without considering backwards compatibility constraints, which are one of the biggest maintenance burdens.
CodesInChaos··on Solving the 1950 cipher meant to test life after death
So we now have evidence that AI absorbs the souls of dead people.
CodesInChaos··on PHP 8.2.35
There is little difference between a tag and a branch in git. There is just the convention that tags reference the same commit forever, while branches move.

On github forks are part of the same physical repository. You could imagine that each logical repository is just a namespace for branches and tags, backed by the same physical repository.

I do consider the warning a bug if github, it should check both tags and branches for commits which are to be considered part of a logical repository.

CodesInChaos··on PHP 8.2.35
I'm surprised by CGI. The vulnerability description talks about a second request being affected by corrupted state from an earlier request. But I thought CGI only processes one request per process?
CodesInChaos··on SlopCodeBench 10/2026 – Opus 5.5, Astra wins overall, xhigh considered harmful
Matches my subjective experience where Opus Extra likes starting sub-agents that burn tokens for little benefit. While High just does the work itself.
CodesInChaos··on Voxtral 3
I think the original SubSync was the useless linear tool I tried last time. Though the Sefzz15 fork added some kind of "piecewise" feature which might work.
CodesInChaos··on Voxtral 3
Though they do have the money to generate terrible TTS audio for many videos. And since google can't fathom a person knowing a foreign language, use those automatically without warning.
CodesInChaos··on Voxtral 3
Considering the progress on speech recognition, why do YouTube subtitles have little in common with what's spoken in the video? Shouldn't Google be able to do a good job at auch a task?
CodesInChaos··on Voxtral 3
Slightly off-topic, but can anybody recommend a local model/tool that fixes timestamps of existing srt subtitles? That should be easier than generating them from scratch.

When I last looked for this a few years ago, I only found a really primitive one that did a linear transformation of all timestamps in the video, and none to correct each line individually.

edit: Claude recommended alass. Any experiences with that one, or better options?

CodesInChaos··on Verisign .com Domain Name Registry Consumer Class Action
That was the idea behind namecoin. But they screwed up their pricing and died.
CodesInChaos··on There's never been a better time to go into math (by: 3blue1brown) [video]
The AI of today is just a tool, albeit a very powerful one. But I don't think it's a big step for it to reach a level where the majority of humans will be unable to to meaningfully contribute to the economy ten years from now.
CodesInChaos··on JPEG XL vs. AVIF for web use-cases
But why does missing JXL support make the AVIF too small? It ends one pixel before the gray background.
CodesInChaos··on JPEG XL vs. AVIF for web use-cases
https://jakearchibald.com/2026/jpeg-xl-vs-avif/#lossy---scre...

In full screen mode, the "hold for lossless" version "Firefox help"/"Submit feedback" are not underlined. In inline mode they are. Both modes seem to render a scaled down version of the image, which seems to cause these problems.

CodesInChaos··on JPEG XL vs. AVIF for web use-cases
Lepton guarantees that JPEG->Lepton->JPEG produces an identical file, not just identical pixels. Does JPEG XL achieve the same thing?
CodesInChaos··on JPEG XL vs. AVIF for web use-cases
In Firefox 157.0.1, the right border of the image moves by a pixel or two when I press "hold for lossless" while viewing the AVIF. I think it's the lossy AVIF that's a bit too small.
CodesInChaos··on Wanted to download an ISO, ended up writing a kernel exploit
TLS over TCP assumes TCP supplies a reliable uncorrupted stream of data. So it has no retries of its own, and terminates the connection when invalid data arrives.

I assume DTLS and QUIC handle MAC verification failures gracefully, like ordinary packet loss.

CodesInChaos··on Poisoned Google SERPs
It's not clear to me if this is actually poisoning, or just some kind of A/B test. Especially since all the "normal" results are SEO slop.
CodesInChaos··on Mozilla Introduces Firefox Enterprise
No public price information. I wonder how much they charge.
CodesInChaos··on Postgres isn't slow, your storage is
The parent post emphasizes how portable schemas or queries cost performance. Now I understand important parts of the schema are postgres specific (e.g. the configuration of non-trivial indexes), but I think most of my queries happen to be portable. So I wondered where people sacrifice performance in the name of portability, and why that's so common that it's worth calling out.
CodesInChaos··on Postgres isn't slow, your storage is
Having indexes fitting the query is of course important, but it's not postgres specific.
CodesInChaos··on Postgres isn't slow, your storage is
I'd like clouds to offer NVMe with value-add features like atomic snapshots and asynchronous replication on top.
Page 1 of 34Next →